No, this is more like any modern processor, which translates instruction codes into micro-ops. To over-simplify IBM just has two of these units per thread rather than one.

They were pretty clear at Hot Chips that it's native

Transmeta did code translation with a software JIT. The hardware side had specific support for x86 semantics, similar to how Apple's M-series has supporting hardware for Rosetta's AOT translation, but neither executes x86 machine code directly.